Based upon results of both our clinical trial of IGFIR antibody treatment in Ewing's sarcoma and rhabdomyosarcoma patients showing rapid development of acquired resistance among responsive patients with most responding patients developing progressive disease within 5 months of treatment, we have focused on studying mechanisms of resistance in the laboratory. We have previously published data showing that rhabdomyosarcoma mouse xenografts develop similar resistance to IGFIR antibody treatment within 60 days. We have now demonstrated that at least 2 mechanisms of resistance are observed. First we found the Src family kinase (SFK) member YES is rapidly activated in rhabdomyosarcoma cell lines and xenografts after exposure to IGFIR antibody. Second we showed that YES is highly expressed in a subset of rhabdomyosarcoma human tumors. Third we showed that simultaneous blockade of SFK with dasatinib or saracatinib leads to enhanced tumor inhibition when combined with IGFIR blockade. These data were recently published. Based upon these published data, we have now been given to approval to open a study testing the combination of the SFK inhibitor, dasatinib with an IGFIR humanized Ab and are currently writing this protocol. We have also established cell lines resistant to IGFIR antibody treatment and used these cell lines to look for phospho-proteomic differences between the resistant and sensitive parental cell lines. We identified PDGFR-beta as being markedly up-regulated in the resistant cell line and have confirmed that concomitant inhibition of PDGFR and IGFIR leads to enhanced cell killing of these cell lines in vitro. Ongoing experiments seek to determine whether this synergy will be seen using human xenograft studies in immundeficient mice. We have completed a kinome siRNA screen in Ewing's sarcoma cells in collaboration with Natasha Caplen. We have identified 11 candidate kinases that appear to be necessary for Ewing's sarcoma cell growth including Aurora kinase B, Chek1, and Polo-like kinase. Of further interest, the importance of several of these kinases were confirmed in an independent high-throughput compound screen against 63 human sarcoma cell lines including 19 Ewing's sarcoma cell lines, in collaboration with Dr. Bev Teicher's group. This screen independently demonstrated that Ewing's sarcoma cells were particularly sensitive to inhibitors of Aurora kinase, Chek1/2, and Polo-like kinase. We are currently focusing on understanding the mechanism of sensitivity to Aurora kinase inhibitors and are testing these inhibitors in xenograft studies. Several of our Ewing's sarcoma xenografts are highly sensitive to Barasertib, an Aurora kinase B inhibitor and we are now trying to identify molecular predictors of high sensitivity. Our whole genome siRNA screen in Ewing's sarcoma cells identified RNA splicing factors as a specific vulnerability in these cells. We have gone on to show that 3 splicing factors, HNRNPH1, SF3B1, and SUPT6H are critical for formation of a mature EWS-FLI-1 fusion product and that inhibition of these factors leads to suppression of Ewing's sarcoma cell growth. We are actively pursuing additional studies of splicing factor inhibitors in Ewing's sarcoma and have recently submitted a manuscript reporting on these findings. We have now fully characterized 95 patients with GIST tumors lacking canonical KIT or PDGFRA mutations. We found 63 patients with mutations in SDHA, SDHB, SDHC or SDHD genes, most of whom have germline mutations in the same gene. An additional 21 patients were found to have a specific methylation of their SDHC promoter leading to silencing of SDHC expression. This combined group of 74 patients are now referred to as SDH deficient GIST patients and have specific clinical features that is leading to novel approaches to treatment. Finally, the remaining 11 patients were found to have intact SDH signaling and we identifed unique mutations in NF1, BRAF, CBL, and a cryptic fusion between KIT and PDGFRA in 9/11 of this group of patients, who behave more similarly to standard KIT/PDGFRA mutant GIST. These data have been published in several papers and another paper is currently in review.